N-Methyl-D-aspartate receptors (NMDARs) are important both in normal and abnormal functioning of the retina. This fact imposes important constraints on possible therapeutic strategies aimed at ameliorating NMDAR-mediated insults to retinal ganglion cells (RGCs), such as ischemia and glaucoma. Block of NMDAR overactivity must be achieved without interference of normal function. Our group was the first to show that the uncompetitive, open-channel blocker, Memantine, can curtail excessive activity of the NMDAR while leaving normal neurotransmission essentially unabated. Based in part on our work, Memantine is being clinically tested as a neuroprotectant for glaucoma and retinal hypoxia/ischemia, and was recently approved for Alzheimer's disease in the USA. Here we develop dual-functional derivatives, the NO-Memantines, that use the Memantine moiety to target NO to S-nitrosylation site(s) on the NMDAR of RGCs to further downregulate excessive activity better than Memantine alone. In our preliminary results, we have shown: (i) the gln/arg/asn (Q/R/N) sites in the second transmembrane (M2) domains of NR1 and NR2 subunits form a specific binding site for Memantine;the binding of Memantine to this site can be used for targeting NO to the NMDAR nitrosylation site;(ii) hypoxia enhances downregulation of NMDAR activity via S-nitrosylation;(Hi) a novel family of NMDAR subunits, termed NR3, exist in the retina (we have cloned and characterized these subunits). Co-expression of NR1/NR2A/NR3A (or 3B) in recombinant systems decreases unitary conductance, Ca2+ permeability, and Mg2+ sensitivity in single-channel recordings of NMDA/glycine-activated currents. Decreased NMDAR activity engendered by NR3 may be protective to RGCs;(iv) co-expression of NR3 and NR1 subunits (without NR2) in Xenopus oocytes and mammalian HEK 293 cells form excitatory glycine-activated cation channels (in the absence of glutamate);(v) S-nitrosylation of the NR1 structure in crystallography experiments. This finding allows us to begin to relate structure to function. We propose the following Specific Aims: 1. To prove the specificity of NO-Memantine targeting to the NMDAR, and to elucidate the mechanism of enhanced downregulation of NMDAR activity by S-nitrosylation under hypoxic conditions. Importantly, the NO-Memantine drugs will also be tested for neuroprotection of RGCs from NMDA or hypoxic-ischemic insults; 2. To characterize the channel pore region of NR3-containing channels; 3. To characterize the ligand-binding site of NR3-containing receptors, and use information from [2] and [3] to distinguish expression of NR1/NR3 "doublet" receptors from NR1/NR2/NR3 "triplet" receptors on RGCs.